Machine Learning for Dynamical Systems

is a center of excellence advancing artificial intelligence and astrodynamics research. By designing state-of-the-art machine learning models, we enable next-generation spacecraft autonomy in complex dynamical environments, and open-source our findings to expedite the rate of adoption within the community.

View Our Projects

Core Research Themes


Astrodynamics


Designing models and algorithms for more sophisticated satellite guidance, navigation, and control.

Scientific Machine Learning


Fuse dynamical systems theory and machine learning to uncover more compact basis functions of complex phenomena.

Spacecraft Autonomy


Enable enhanced spacecraft autonomy by leveraging partially observable Markov decision processes and deep reinforcement learning.

John Martin

John Martin

Assistant Professor of Aerospace Engineering

University of Maryland

Biography

John Martin is an assistant professor of aerospace engineering at the University of Maryland. His research interests include astrodynamics and scientific machine learning. He leads the Machine Learning for Dynamical Systems (MLDS) group, which develops open-source machine learning dynamics models for use in spacecraft guidance, navigation, control, and planning.

Interests
  • Astrodynamics
  • Scientific Machine Learning
  • Spacecraft Autonomy
  • Software Engineering
Education
  • PhD Aerospace Engineering, 2023

    University of Colorado Boulder

  • MS in Aerospace Engineering, 2021

    University of Colorado Boulder

  • BSc in Physics and Astronomy, 2018

    University of North Carolina at Chapel Hill